#463 by GitHub stars
amElnagdy/delegate-skills
Delegate a coding task to a separate coding agent CLI, review the diff, land the commit yourself — one per implementer.
History
Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.
View weekly changes
skills/agy-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/aider-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/claude-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/cline-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/codex-background/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/codex-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/commandcode-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/copilot-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/cursor-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/delegate-setup/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/grok-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/kimi-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/muse-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/omp-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/opencode-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/pi-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/qoder-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/vibe-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/warp-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
skills/zcode-delegate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
3aa72d681a79 → 3aa72d681a79
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 3aa72d681a79
No earlier version was observed for comparison.
View on GitHubAdd to my skills
Verified commit 3aa72d681a79e2a9d09bd9a316353003ae1cb4a7
git clone --filter=blob:none --no-checkout https://github.com/amElnagdy/delegate-skills.git .hotskills/amelnagdy-delegate-skills && git -C .hotskills/amelnagdy-delegate-skills checkout --detach 3aa72d681a79e2a9d09bd9a316353003ae1cb4a7
# Copy each selected skill directory from .hotskills/amelnagdy-delegate-skills into .agents/skills/<skill-name>/git clone --filter=blob:none --no-checkout https://github.com/amElnagdy/delegate-skills.git .hotskills/amelnagdy-delegate-skills && git -C .hotskills/amelnagdy-delegate-skills checkout --detach 3aa72d681a79e2a9d09bd9a316353003ae1cb4a7
# Copy each selected skill directory from .hotskills/amelnagdy-delegate-skills into .cursor/skills/<skill-name>/2.6kGitHub stars
28stars in 7 days
20skills in this repository
46commits in 30 days
Heat
6.8- Star growth, last 7 daysAhead of 86% of tracked repositories
- Star growth, last 24 hoursAhead of 90% of tracked repositories
- Rank movement, last 7 daysAhead of 40% of tracked repositories
- Commits, last 30 daysAhead of 90% of tracked repositories
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekAhead of 0% of tracked repositories
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
Individual skills in this repository
- agy-delegate
skills/agy-delegate - aider-delegate
skills/aider-delegate - claude-delegate
skills/claude-delegate - cline-delegate
skills/cline-delegate - codex-background
skills/codex-background - codex-delegate
skills/codex-delegate - commandcode-delegate
skills/commandcode-delegate - copilot-delegate
skills/copilot-delegate - cursor-delegate
skills/cursor-delegate - delegate-setup
skills/delegate-setup - grok-delegate
skills/grok-delegate - kimi-delegate
skills/kimi-delegate - muse-delegate
skills/muse-delegate - omp-delegate
skills/omp-delegate - opencode-delegate
skills/opencode-delegate - pi-delegate
skills/pi-delegate - qoder-delegate
skills/qoder-delegate - vibe-delegate
skills/vibe-delegate - warp-delegate
skills/warp-delegate - zcode-delegate
skills/zcode-delegate
How it works
agy-delegate
When your agent uses it
Delegate a coding task to the Google Antigravity CLI (`agy`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to Antigravity or agy - phrasings like "have Antigravity do X", "delegate this to agy", "run it through agy", or "use Antigravity to implement/fix/refactor" - or wants to run a queue of coding tasks through agy while staying the reviewer. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating.
What it loads into context
Frontmatter
218 tokensAlways available to the agent
SKILL.md
2,583 tokensWhen the skill is activated
Reference documents
7,291 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Prerequisites (check once)
- Choose the implementer model
- The loop
- Permission model
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore10,829 bytesSKILL.mdcore10,829 bytesSKILL.mdcore10,829 bytesreferences/dispatch-and-poll.mddoc16,265 bytesreferences/multi-task-queues.mddoc2,960 bytesreferences/review-and-land.mddoc5,408 bytesreferences/writing-the-brief.mddoc5,882 bytesscripts/relay.mjsruns41,187 bytes
Read SKILL.md
# Antigravity Delegate
You are the **orchestrator**. This skill lets you hand a bounded coding task to a separate
**implementer** - the Google Antigravity CLI (`agy`) - then review what it produced and land it
yourself. You write the brief and own the judgment; Antigravity does the typing in its own
conversation; you verify and commit.
Nothing here is specific to one orchestrating agent. The loop needs only the ability to run a shell
command and read a file, so any comparable agent can drive it. It is designed for and run on Claude
Code; treat other orchestrators as designed-for, not yet proven.
## When NOT to use this
- The task is small enough to just do inline - delegation overhead is not worth it.aider-delegate
When your agent uses it
Delegate a coding task to Aider (`aider`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to Aider - phrasings like "have Aider do X", "delegate this to aider", "run it through Aider", or "use Aider to implement/fix/refactor" - or wants to run a queue of coding tasks through Aider while staying the reviewer. This includes asking Aider to drive a local or self-hosted OpenAI-compatible endpoint ("have Aider use my local model", "run Aider against llama.cpp / Ollama / vLLM / LM Studio"), which Aider reaches via `--api-base`. DO NOT USE for local-model or coding requests that do not name Aider, for tasks small enough to do inline, or when the user wants the code written directly without delegating.
What it loads into context
Frontmatter
351 tokensAlways available to the agent
SKILL.md
3,120 tokensWhen the skill is activated
Reference documents
5,618 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The one thing to know about Aider
- When NOT to use this
- Prerequisites (check once)
- Choose the model
- Local and self-hosted models
- The loop
- Autonomy and permissions
- Resume
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore12,911 bytesSKILL.mdcore12,911 bytesSKILL.mdcore12,911 bytesSKILL.mdcore12,911 bytesreferences/dispatch-and-poll.mddoc9,182 bytesreferences/multi-task-queues.mddoc2,868 bytesreferences/review-and-land.mddoc5,108 bytesreferences/writing-the-brief.mddoc5,681 bytesscripts/relay.mjsruns35,091 bytes
Read SKILL.md
# Aider Delegate
You are the **orchestrator**. Hand a bounded coding task to a separate **implementer** - Aider - then
review what it produced and land it yourself. You write the brief and own the judgment; Aider does the
typing in its own run; you verify and commit.
The loop needs only a shell command and file access, so any comparable orchestrator can drive it.
## The one thing to know about Aider
**Aider commits by default.** Two of its defaults would destroy the reviewable diff this skill exists
to produce:
- `--auto-commits` (default `True`) - Aider commits its own edits after each exchange.claude-delegate
When your agent uses it
Delegate a coding task to a separate Claude Code CLI process or another Claude session as an implementer, then review its diff and land it yourself. Use only when the user explicitly asks to delegate implementation to Claude Code, another Claude session, or the `claude` CLI — for example, "have another Claude implement this", "delegate this to Claude Code", or "run this queue through a separate Claude session." Do not trigger merely because the current orchestrator is Claude, and do not use when the user asks the current Claude to implement directly without delegation.
What it loads into context
Frontmatter
249 tokensAlways available to the agent
SKILL.md
2,335 tokensWhen the skill is activated
Reference documents
6,406 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When not to use this
- Prerequisites
- The loop
- Permission profiles
- Complementary to native Claude features
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore10,478 bytesSKILL.mdcore10,478 bytesSKILL.mdcore10,478 bytesSKILL.mdcore10,478 bytesSKILL.mdcore10,478 bytesreferences/dispatch-and-poll.mddoc13,309 bytesreferences/multi-task-queues.mddoc2,814 bytesreferences/review-and-land.mddoc5,492 bytesreferences/writing-the-brief.mddoc7,121 bytesscripts/relay.mjsruns49,760 bytes
Read SKILL.md
# Claude Delegate
You are the **orchestrator**. Delegate one bounded coding task to a separate **implementer** — a Claude
Code CLI session — then review what it produced and land it yourself. You write the brief and own the
judgment; the separate Claude session edits the working tree; you verify and commit.
This skill is not a signal for the current Claude to implement directly. Use it only after the human
explicitly asks for delegation to another Claude Code process or session.
## When not to use this
- The human asked the current agent to implement the task directly.
- The task is small enough to do inline and the human did not request delegation.
- The `claude` CLI is missing or unauthenticated (`claude auth status`).cline-delegate
When your agent uses it
Delegate a coding task to the Cline coding agent CLI (`cline`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to delegate implementation work to Cline - phrasings like "have Cline implement X", "delegate this to cline", "run it through Cline", or "use cline to implement/fix/refactor" - or wants to run a queue of coding tasks through Cline while staying the reviewer. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating.
What it loads into context
Frontmatter
197 tokensAlways available to the agent
SKILL.md
1,676 tokensWhen the skill is activated
Reference documents
4,681 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Prerequisites (check once)
- Choose the model (optional)
- The loop
- Autonomy and permissions
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore7,048 bytesSKILL.mdcore7,048 bytesSKILL.mdcore7,048 bytesreferences/dispatch-and-poll.mddoc8,281 bytesreferences/multi-task-queues.mddoc2,280 bytesreferences/review-and-land.mddoc3,654 bytesreferences/writing-the-brief.mddoc5,583 bytesscripts/relay.mjsruns32,674 bytes
Read SKILL.md
# Cline Delegate
You are the **orchestrator**. Delegate a bounded coding task to a separate **implementer** - the Cline
coding agent CLI - then review what it produced and land it yourself. You write the brief and own
the judgment; the implementer makes changes in its own session in a clean working tree; you verify
and commit.
The loop needs only a shell command and file access, so any comparable orchestrator can drive it.
## When NOT to use this
- The task is small enough to do inline; delegation overhead is not worth it.
- The `cline` CLI is not installed or authenticated.
- You require the relay to configure a sandbox. Cline exposes sandbox controls, but this relaycodex-background
When your agent uses it
Keep a Codex orchestrator's delegation to an installed delegate skill pending until its registered relay completes, without model-driven progress polling. Use with a selected implementer delegate skill when Codex is the orchestrator and the delegation should wait in the background; sets up its MCP server on first use.
What it loads into context
Frontmatter
139 tokensAlways available to the agent
SKILL.md
867 tokensWhen the skill is activated
Reference documents
4,372 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- First use / setup
- Procedure
Scripts executed separately
scripts/bootstrap.mjsscripts/bootstrap.test.mjsscripts/server.mjsscripts/server.test.mjsscripts/supervisor.mjsscripts/windows-job.ps1
Files
SKILL.mdcore4,207 bytesSKILL.mdcore4,207 bytesSKILL.mdcore4,207 bytesreferences/configuration.mddoc20,269 bytesscripts/bootstrap.mjsruns36,693 bytesscripts/bootstrap.test.mjsruns21,400 bytesscripts/linux-owner.casset4,760 bytesscripts/server.mjsruns20,931 bytesscripts/server.test.mjsruns23,507 bytesscripts/supervisor.mjsruns3,208 bytesscripts/windows-job.csasset5,855 bytesscripts/windows-job.ps1runs554 bytes
Read SKILL.md
# Codex background delegation
This support skill changes Codex's waiting boundary. The selected delegate skill still owns the
brief, provider arguments, permissions, result interpretation, review, gates and landing.
Claude's background Bash workflow stays unchanged. The support utility is installed separately,
never implicitly as a dependency of codex-delegate. Details: [references/configuration.md](references/configuration.md).
## First use / setup
If the `delegate_run` tool is not available, setup is incomplete. Tell the user that bootstrap will
update `$CODEX_HOME/config.toml` (default `~/.codex/config.toml`, backed up first) and write a registry of
the installed `*-delegate` skills, then run:
```bashcodex-delegate
When your agent uses it
Delegate a coding task to the OpenAI Codex CLI as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to Codex — phrasings like "have Codex do X", "delegate this to Codex", "run it through Codex", or "use Codex to implement/fix/refactor" — or to run a queue of coding tasks through Codex while staying the reviewer. Prefer it over a one-shot Codex forwarder (such as the codex-rescue agent) when the user will review the diff and commit it themselves. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating.
What it loads into context
Frontmatter
250 tokensAlways available to the agent
SKILL.md
2,810 tokensWhen the skill is activated
Reference documents
12,904 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Prerequisites (check once)
- The loop
- Read-only second opinions
- Authorization model
- If you have the openai-codex plugin
- References
Scripts executed separately
scripts/relay.mjs
Files
../codex-background/references/configuration.mddoc20,269 bytesSKILL.mdcore11,694 bytesSKILL.mdcore11,694 bytesSKILL.mdcore11,694 bytesreferences/dispatch-and-poll.mddoc18,209 bytesreferences/multi-task-queues.mddoc3,741 bytesreferences/review-and-land.mddoc7,877 bytesreferences/writing-the-brief.mddoc6,072 bytesscripts/relay.mjsruns39,129 bytes
Read SKILL.md
# Codex Delegate
You are the **orchestrator**. This skill lets you hand a bounded coding task to a separate
**implementer** — the OpenAI Codex CLI — then review what it produced and land it yourself. You write
the brief and own the judgment; Codex does the typing in its own sandbox; you verify and commit.
Nothing here is specific to one orchestrating agent. The loop needs only the ability to run a shell
command and read a file, so it works the same whether you are Claude Code, OpenCode with a selected
model, or any comparable agent. (It is designed for and run on Claude Code; treat other orchestrators
as designed-for, not yet proven.)
## When NOT to use this
- The task is small enough to just do inline — delegation overhead is not worth it.commandcode-delegate
When your agent uses it
Delegate a coding task to the Command Code CLI (`cmd`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to Command Code — phrasings like "have Command Code do X", "delegate this to commandcode", "run it through cmd", or "use Command Code to implement/fix/refactor" — or to run a queue of coding tasks through Command Code while staying the reviewer. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating.
What it loads into context
Frontmatter
216 tokensAlways available to the agent
SKILL.md
2,704 tokensWhen the skill is activated
Reference documents
9,068 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Read this before the first dispatch: the autonomy model
- Prerequisites (check once)
- The loop
- Read-only second opinions
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore11,454 bytesSKILL.mdcore11,454 bytesSKILL.mdcore11,454 bytesreferences/dispatch-and-poll.mddoc16,456 bytesreferences/multi-task-queues.mddoc4,840 bytesreferences/review-and-land.mddoc9,632 bytesreferences/writing-the-brief.mddoc7,599 bytesscripts/relay.mjsruns58,594 bytes
Read SKILL.md
# Command Code Delegate
You are the **orchestrator**. This skill lets you hand a bounded coding task to a separate
**implementer** — the Command Code CLI (`cmd`) — then review what it produced and land it yourself.
You write the brief and own the judgment; Command Code does the typing in your working tree; you
verify and commit.
Nothing here is specific to one orchestrating agent. The loop needs only the ability to run a shell
command and read a file, so it works the same whether you are Claude Code, OpenCode with a selected
model, or any comparable agent. (It is designed for and run on Claude Code; treat other orchestrators
as designed-for, not yet proven.)
## When NOT to use this
copilot-delegate
When your agent uses it
Delegate a coding task to the GitHub Copilot CLI (`copilot`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to delegate implementation work to Copilot - phrasings like "have Copilot implement X", "delegate this to copilot", "run it through Copilot CLI", or "use copilot to implement/fix/refactor" - or wants to run a queue of coding tasks through Copilot while staying the reviewer. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating.
What it loads into context
Frontmatter
217 tokensAlways available to the agent
SKILL.md
1,874 tokensWhen the skill is activated
Reference documents
4,849 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Prerequisites (check once)
- Choose the model (optional)
- Choose the effort (optional)
- The loop
- Autonomy and permissions
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore7,974 bytesSKILL.mdcore7,974 bytesSKILL.mdcore7,974 bytesreferences/dispatch-and-poll.mddoc8,060 bytesreferences/multi-task-queues.mddoc2,280 bytesreferences/review-and-land.mddoc3,671 bytesreferences/writing-the-brief.mddoc6,399 bytesscripts/relay.mjsruns31,578 bytes
Read SKILL.md
# Copilot Delegate
You are the **orchestrator**. Delegate a bounded coding task to a separate **implementer** — the
GitHub Copilot CLI — then review what it produced and land it yourself. You write the brief and own
the judgment; the implementer makes changes in its own session in a clean working tree; you verify
and commit.
The loop needs only a shell command and file access, so any comparable orchestrator can drive it.
## When NOT to use this
- The task is small enough to do inline; delegation overhead is not worth it.
- The `copilot` CLI is not installed or authenticated.
- You need a hard sandbox. Copilot exposes sandbox controls, but they are upstream-experimentalcursor-delegate
When your agent uses it
Delegate a coding task to the Cursor Agent CLI (`cursor-agent`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to Cursor — phrasings like "have Cursor implement X", "delegate this to Cursor", "run it through Cursor Agent", or "use Cursor to implement/fix/refactor" — or wants to run a queue of coding tasks through Cursor while staying the reviewer. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating.
What it loads into context
Frontmatter
229 tokensAlways available to the agent
SKILL.md
2,191 tokensWhen the skill is activated
Reference documents
5,339 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Prerequisites (check once)
- Choose the model
- The loop
- Autonomy and permissions
- Read-only second opinions
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore9,503 bytesSKILL.mdcore9,503 bytesSKILL.mdcore9,503 bytesreferences/dispatch-and-poll.mddoc10,527 bytesreferences/multi-task-queues.mddoc2,326 bytesreferences/review-and-land.mddoc4,552 bytesreferences/writing-the-brief.mddoc5,316 bytesscripts/relay.mjsruns35,081 bytes
Read SKILL.md
# Cursor Delegate
You are the **orchestrator**. Hand a bounded coding task to a separate **implementer** — the Cursor
Agent CLI — then review what it produced and land it yourself. You write the brief and own the
judgment; Cursor does the typing in its own session; you verify and commit.
The loop needs only a shell command and file access, so any comparable orchestrator can drive it.
## When NOT to use this
- The task is small enough to do inline; delegation overhead is not worth it.
- The `cursor-agent` CLI is not installed or authenticated (run `cursor-agent login`).
- You want to write the code yourself, or you only need Cursor's opinion on code you wrote (a
`--read-only` dispatch covers that — see below — but a plain review may not need delegation at all).delegate-setup
When your agent uses it
Configure delegation fleet lanes: which implementer CLI handles which kind of work, with optional model and effort (or variant) dials. Discovers installed CLIs, proposes a lane map for user approval, and writes global or project config only after explicit yes. Use when the user asks to set up, configure, or reconfigure delegation lanes, a fleet of lanes, or which implementer handles feature/tests/ui work — not for dispatching a coding task to an implementer.
What it loads into context
Frontmatter
153 tokensAlways available to the agent
SKILL.md
2,588 tokensWhen the skill is activated
Reference documents
3,231 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Hard rules
- Flow
- Reconfigure
Scripts executed separately
scripts/config.mjsscripts/discover.mjsscripts/implementers.mjsscripts/lane.mjs
Files
SKILL.mdcore10,929 bytesSKILL.mdcore10,929 bytesSKILL.mdcore10,929 bytesSKILL.mdcore10,929 bytesSKILL.mdcore10,929 bytesSKILL.mdcore10,929 bytesSKILL.mdcore10,929 bytesreferences/schema.mddoc5,415 bytesreferences/setup-dialogue.mddoc6,990 bytesscripts/config.mjsruns18,457 bytesscripts/discover.mjsruns16,407 bytesscripts/implementers.mjsruns17,538 bytesscripts/lane.mjsruns6,817 bytes
Read SKILL.md
# Delegate Setup
You are the **orchestrator** in **setup mode**. Discover installed implementer CLIs, propose a
**fleet of lanes**, and write configuration only after the user approves.
This skill does **not** dispatch coding work. It only authors the lane map.
One concept: **lanes**. Never say “routes.”
Example lane: **feature** → implementer `opencode`, model `opencode/grok`, variant `high`
(OpenCode uses `variant` for reasoning intensity, not `effort`).
## When NOT to use this
grok-delegate
When your agent uses it
Delegate a coding task to the Grok Build CLI as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to Grok — phrasings like "have Grok do X", "delegate this to Grok", "run it through Grok", "use Grok Build to implement/fix/refactor", or "have grok CLI do this" — or to run a queue of coding tasks through Grok while staying the reviewer. Prefer it when the user will review the diff and commit it themselves. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating.
What it loads into context
Frontmatter
260 tokensAlways available to the agent
SKILL.md
2,721 tokensWhen the skill is activated
Reference documents
7,126 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Prerequisites (check once)
- The loop
- Autonomy model
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore11,241 bytesSKILL.mdcore11,241 bytesSKILL.mdcore11,241 bytesreferences/dispatch-and-poll.mddoc12,578 bytesreferences/multi-task-queues.mddoc3,841 bytesreferences/review-and-land.mddoc8,136 bytesreferences/writing-the-brief.mddoc5,772 bytesscripts/relay.mjsruns53,485 bytes
Read SKILL.md
# Grok Delegate
For a trusted repository rejected by Git's ownership check, the relay supports
`--trust-git-root <exact-worktree-root>`. This opt-in affects only relay Git checks, without persistent
Git config or Grok permission changes. See [dispatch and poll](references/dispatch-and-poll.md#git-ownership-errors-on-shared-or-remounted-drives).
You are the **orchestrator**. This skill lets you hand a bounded coding task to a separate
**implementer** — the Grok Build CLI (`grok`) — then review what it produced and land it yourself. You
write the brief and own the judgment; Grok does the typing under an explicit autonomy profile; you
verify and commit.
Nothing here is specific to one orchestrating agent. The loop needs only the ability to run a shell
command and read a file, so it works the same whether you are Claude Code, Cursor, OpenCode with a
selected model, or any comparable agent. (It is designed for Claude Code and Cursor; treat otherkimi-delegate
When your agent uses it
Delegate a coding task to the Kimi Code CLI (`kimi`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to Kimi - phrasings like "have Kimi implement X", "delegate this to Kimi", "run it through Kimi Code", or "use Kimi to implement/fix/refactor" - or wants to run a queue of coding tasks through Kimi while staying the reviewer. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating.
What it loads into context
Frontmatter
157 tokensAlways available to the agent
SKILL.md
1,657 tokensWhen the skill is activated
Reference documents
4,677 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Prerequisites (check once)
- Choose the model alias
- The loop
- Autonomy and permissions
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore6,983 bytesSKILL.mdcore6,983 bytesSKILL.mdcore6,983 bytesreferences/dispatch-and-poll.mddoc7,451 bytesreferences/multi-task-queues.mddoc2,314 bytesreferences/review-and-land.mddoc4,478 bytesreferences/writing-the-brief.mddoc5,588 bytesscripts/relay.mjsruns26,548 bytes
Read SKILL.md
# Kimi Delegate
You are the **orchestrator**. Hand a bounded coding task to a separate **implementer** - the Kimi Code
CLI - then review what it produced and land it yourself. You write the brief and own the judgment;
Kimi does the typing in its own session; you verify and commit.
The loop needs only a shell command and file access, so any comparable orchestrator can drive it.
## When NOT to use this
- The task is small enough to do inline; delegation overhead is not worth it.
- The `kimi` CLI is not installed or authenticated.
- You need a CLI-enforced read-only implementer. Headless Kimi has no read-only mode.
muse-delegate
When your agent uses it
Delegate a coding task to Muse Code (`muse`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to Muse — phrasings like "have Muse do X", "delegate this to Muse", "run it through Muse", or "use Muse to implement/fix/refactor" — or wants to run a queue of coding tasks through Muse while staying the reviewer. Prefer it when the user will review the diff and commit it themselves. It applies only when the user wants that separate `muse exec` implementer run: if you are Muse yourself and the user simply asks you to write the code, do it directly. DO NOT USE for tasks small enough to do inline.
What it loads into context
Frontmatter
257 tokensAlways available to the agent
SKILL.md
2,325 tokensWhen the skill is activated
Reference documents
6,659 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Read this before the first dispatch: the autonomy model
- Prerequisites (check once)
- Choose the implementer model
- The loop
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore9,489 bytesSKILL.mdcore9,489 bytesSKILL.mdcore9,489 bytesreferences/dispatch-and-poll.mddoc9,493 bytesreferences/multi-task-queues.mddoc4,025 bytesreferences/review-and-land.mddoc7,884 bytesreferences/writing-the-brief.mddoc6,703 bytesscripts/relay.mjsruns27,128 bytes
Read SKILL.md
# Muse Delegate
You are the **orchestrator**. This skill lets you hand a bounded coding task to a separate
**implementer** — Muse Code (`muse`) — then review what it produced and land it yourself. You write
the brief and own the judgment; Muse does the typing in its own `exec` run; you verify and commit.
Nothing here is specific to one orchestrating agent. The loop needs only the ability to run a shell
command and read a file, so any agent with those two capabilities can drive it. The full delegate →
review → commit loop has been run with Claude Code as the orchestrator. Other orchestrators remain
unproven.
## When NOT to use this
- The task is small enough to just do inline — delegation overhead is not worth it.omp-delegate
When your agent uses it
Delegate a coding task to Oh My Pi (`omp`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to delegate implementation work to Oh My Pi / omp - phrasings like "have omp implement X", "delegate this to oh my pi", "run it through omp", "use oh-my-pi to implement/fix/refactor" - or wants to run a queue of coding tasks through omp while staying the reviewer. DO NOT USE for tasks small enough to do inline, when the user wants the code written directly without delegating, or when they mean the original Pi CLI (`pi`) — that is pi-delegate.
What it loads into context
Frontmatter
244 tokensAlways available to the agent
SKILL.md
2,401 tokensWhen the skill is activated
Reference documents
5,444 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The binary is `omp`, not `pi`
- When NOT to use this
- Prerequisites (check once)
- Choose the model (optional)
- The loop
- Autonomy and permissions
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore9,892 bytesSKILL.mdcore9,892 bytesSKILL.mdcore9,892 bytesreferences/dispatch-and-poll.mddoc8,996 bytesreferences/multi-task-queues.mddoc2,312 bytesreferences/review-and-land.mddoc4,691 bytesreferences/writing-the-brief.mddoc6,825 bytesscripts/relay.mjsruns32,878 bytes
Read SKILL.md
# Oh My Pi Delegate
You are the **orchestrator**. Delegate a bounded coding task to a separate **implementer** - Oh My
Pi (`omp`) - then review what it produced and land it yourself. You write the brief and own the
judgment; the implementer makes changes in its own session; you verify and commit.
The loop needs only a shell command and file access, so any comparable orchestrator can drive it.
## The binary is `omp`, not `pi`
Oh My Pi is a fork of Pi. This skill drives **`omp`** (`@oh-my-pi/pi-coding-agent`). The original
Pi CLI is a different binary (`pi`) with a different skill (`pi-delegate`). If `omp` is missing but
`pi` is installed, you have Pi, not Oh My Pi.
opencode-delegate
When your agent uses it
Delegate a coding task to the OpenCode CLI as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to OpenCode — phrasings like "have OpenCode do X", "delegate this to OpenCode", "run it through OpenCode", or "use OpenCode to implement/fix/refactor" — or wants to run a queue of coding tasks through OpenCode while staying the reviewer. Prefer it when the user will review the diff and commit it themselves. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating.
What it loads into context
Frontmatter
224 tokensAlways available to the agent
SKILL.md
2,764 tokensWhen the skill is activated
Reference documents
7,122 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Prerequisites (check once)
- Choose the implementer model
- The loop
- Autonomy model
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore11,461 bytesSKILL.mdcore11,461 bytesSKILL.mdcore11,461 bytesSKILL.mdcore11,461 bytesreferences/dispatch-and-poll.mddoc10,911 bytesreferences/multi-task-queues.mddoc3,928 bytesreferences/review-and-land.mddoc7,851 bytesreferences/writing-the-brief.mddoc7,524 bytesscripts/relay.mjsruns34,856 bytes
Read SKILL.md
# OpenCode Delegate
You are the **orchestrator**. This skill lets you hand a bounded coding task to a separate
**implementer** — the OpenCode CLI — then review what it produced and land it yourself. You write
the brief and own the judgment; OpenCode does the typing in its own session; you verify and commit.
Nothing here is specific to one orchestrating agent. The loop needs only the ability to run a shell
command and read a file, so any agent with those two capabilities — Claude Code, OpenCode driving a
sibling session, or a comparable one — can drive it. (It is designed for and run on Claude Code; treat
other orchestrators as designed-for, not yet proven.)
## When NOT to use this
- The task is small enough to just do inline — delegation overhead is not worth it.pi-delegate
When your agent uses it
Delegate a coding task to the Pi coding agent CLI (`pi`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to delegate implementation work to Pi - phrasings like "have Pi implement X", "delegate this to pi", "run it through Pi", or "use pi to implement/fix/refactor" - or wants to run a queue of coding tasks through Pi while staying the reviewer. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating.
What it loads into context
Frontmatter
150 tokensAlways available to the agent
SKILL.md
1,778 tokensWhen the skill is activated
Reference documents
5,017 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Prerequisites (check once)
- Choose the model (optional)
- The loop
- Autonomy and permissions
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore7,541 bytesSKILL.mdcore7,541 bytesSKILL.mdcore7,541 bytesreferences/dispatch-and-poll.mddoc8,386 bytesreferences/multi-task-queues.mddoc2,310 bytesreferences/review-and-land.mddoc4,732 bytesreferences/writing-the-brief.mddoc5,833 bytesscripts/relay.mjsruns32,930 bytes
Read SKILL.md
# Pi Delegate
You are the **orchestrator**. Delegate a bounded coding task to a separate **implementer** - the Pi
coding agent CLI - then review what it produced and land it yourself. You write the brief and own
the judgment; the implementer makes changes in its own session; you verify and commit.
The loop needs only a shell command and file access, so any comparable orchestrator can drive it.
## When NOT to use this
- The task is small enough to do inline; delegation overhead is not worth it.
- The `pi` CLI is not installed or authenticated.
- You need a sandboxed implementer. Pi has no sandbox and no permission modes; `--read-only`
restricts the tool surface, but a write-capable run executes without prompts.qoder-delegate
When your agent uses it
Delegate a coding task to the Qoder CLI (`qodercli`) as a background implementer, then review its diff and land it yourself. Use this whenever the user asks to have Qoder implement, fix, refactor, or run a queue of coding tasks while the orchestrator remains the reviewer. DO NOT USE for tasks small enough to do inline, or when the user wants code written directly without delegation.
What it loads into context
Frontmatter
115 tokensAlways available to the agent
SKILL.md
1,615 tokensWhen the skill is activated
Reference documents
3,122 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Prerequisites (check once)
- Choose model and context window
- The loop
- Permission model
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore7,010 bytesSKILL.mdcore7,010 bytesSKILL.mdcore7,010 bytesSKILL.mdcore7,010 bytesreferences/dispatch-and-poll.mddoc5,721 bytesreferences/multi-task-queues.mddoc1,563 bytesreferences/review-and-land.mddoc2,526 bytesreferences/writing-the-brief.mddoc3,469 bytesscripts/relay.mjsruns25,706 bytes
Read SKILL.md
# Qoder Delegate
You are the **orchestrator**. Delegate one bounded coding task to a separate **implementer** - Qoder CLI -
then review what it produced and land it yourself. You write the brief and own the judgment; Qoder
edits the working tree in its session; you verify and commit.
The loop needs only shell and file access, so any comparable orchestrator can drive it.
## When NOT to use this
- The task is small enough to do inline; delegation overhead is not worth it.
- `qodercli` is not installed or authenticated.
- You want to write the code yourself or need only an interactive Qoder session.
vibe-delegate
When your agent uses it
Delegate a coding task to the Mistral Vibe CLI (`vibe`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to Vibe — phrasings like "have Vibe implement X", "delegate this to Vibe", "run it through Mistral Vibe", "use vibe to implement/fix/refactor" — or wants to run a queue of coding tasks through Vibe while staying the reviewer. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating.
What it loads into context
Frontmatter
160 tokensAlways available to the agent
SKILL.md
1,717 tokensWhen the skill is activated
Reference documents
4,733 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Prerequisites (check once)
- The loop
- Autonomy and permissions
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore7,353 bytesSKILL.mdcore7,353 bytesSKILL.mdcore7,353 bytesreferences/dispatch-and-poll.mddoc7,936 bytesreferences/multi-task-queues.mddoc2,371 bytesreferences/review-and-land.mddoc4,075 bytesreferences/writing-the-brief.mddoc5,667 bytesscripts/relay.mjsruns28,184 bytes
Read SKILL.md
# Vibe Delegate
You are the **orchestrator**. Hand a bounded coding task to a separate **implementer** — the Mistral
Vibe CLI (`vibe`) — then review what it produced and land it yourself. You write the brief and own
the judgment; Vibe does the typing in its own session; you verify and commit.
The loop needs only a shell command and file access, so any comparable orchestrator can drive it.
## When NOT to use this
- The task is small enough to do inline; delegation overhead is not worth it.
- The `vibe` CLI is not installed or authenticated.
## Prerequisites (check once)warp-delegate
When your agent uses it
Delegate a coding task to the Warp Agent CLI (`oz`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to Warp - phrasings like "have Warp implement X", "delegate this to the Warp CLI", "run it through Warp", "use oz to implement/fix/refactor" - or wants to run a queue of coding tasks through Warp while staying the reviewer. DO NOT USE for tasks small enough to do inline, when the user wants the code written directly without delegating, or for the interactive `warp` TUI (this skill drives the headless `oz agent run`, not the terminal app).
What it loads into context
Frontmatter
267 tokensAlways available to the agent
SKILL.md
2,665 tokensWhen the skill is activated
Reference documents
6,216 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The binary is `oz`, not `warp`
- When NOT to use this
- Prerequisites (check once)
- Choose the model (optional)
- The loop
- Autonomy and permissions
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore11,095 bytesSKILL.mdcore11,095 bytesSKILL.mdcore11,095 bytesreferences/dispatch-and-poll.mddoc8,651 bytesreferences/multi-task-queues.mddoc3,713 bytesreferences/review-and-land.mddoc7,520 bytesreferences/writing-the-brief.mddoc5,124 bytesscripts/relay.mjsruns33,990 bytes
Read SKILL.md
# Warp Delegate
You are the **orchestrator**. Delegate a bounded coding task to a separate **implementer** - the
Warp Agent CLI - then review what it produced and land it yourself. You write the brief and own the
judgment; the implementer makes changes in its own conversation; you verify and commit.
The loop needs only a shell command and file access, so any comparable orchestrator can drive it.
## The binary is `oz`, not `warp`
Warp ships two different programs, and only one of them can be delegated to:
- **`oz`** - the Warp Agent CLI. Headless and scriptable; `oz agent run` executes an agent against a
local directory. **This is what the relay drives.**zcode-delegate
When your agent uses it
Delegate a coding task to the Z.AI ZCode CLI as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to ZCode — phrasings like "have ZCode do X", "delegate this to ZCode", "run it through ZCode", or "use ZCode to implement/fix/refactor" — or to run a queue of coding tasks through ZCode while staying the reviewer. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating.
What it loads into context
Frontmatter
242 tokensAlways available to the agent
SKILL.md
2,545 tokensWhen the skill is activated
Reference documents
4,776 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When NOT to use this
- Prerequisites (check once)
- Authenticating the headless CLI
- Autonomy — read this before dispatching
- The loop
- Read-only second opinions
- Authorization model
- References
Scripts executed separately
scripts/relay.mjs
Files
SKILL.mdcore10,404 bytesSKILL.mdcore10,404 bytesSKILL.mdcore10,404 bytesreferences/dispatch-and-poll.mddoc6,534 bytesreferences/multi-task-queues.mddoc3,626 bytesreferences/review-and-land.mddoc4,139 bytesreferences/writing-the-brief.mddoc4,968 bytesscripts/relay.mjsruns48,486 bytes
Read SKILL.md
# ZCode Delegate
You are the **orchestrator**. This skill lets you hand a bounded coding task to a separate
**implementer** — the Z.AI ZCode CLI — then review what it produced and land it yourself. You write
the brief and own the judgment; ZCode does the typing; you verify and commit.
Nothing here is specific to one orchestrating agent. The loop needs only the ability to run a shell
command and read a file. (It is designed for and run on Claude Code; treat other orchestrators as
designed-for, not yet proven.)
## When NOT to use this
- The task is small enough to just do inline — delegation overhead is not worth it.
- ZCode is not installed, or its CLI has no model provider configured.Skills included
- agy-delegate
- aider-delegate
- claude-delegate
- cline-delegate
- codex-background
- codex-delegate
- commandcode-delegate
- copilot-delegate
- cursor-delegate
- delegate-setup
- grok-delegate
- kimi-delegate
- muse-delegate
- omp-delegate
- opencode-delegate
- pi-delegate
- qoder-delegate
- vibe-delegate
- warp-delegate
- zcode-delegate
Repository facts
- License
- MIT
- Created
- Jun 14, 2026
- Last pushed
- Nov 5, 2026
- Forks
- 220
- Open issues
- 22
- Project activity
- Active
Rank over the last 12 weeks
Now #463 · best #400 · down 41 in 30 days
- #451 on Oct 28
- #453 on Oct 29
- #455 from Oct 30 to Oct 31
- #458 on Nov 1
- #460 from Nov 2 to Nov 5
- #463 since Nov 6
79 earlier periods
- #695 on Aug 2
- #674 on Aug 7
- #666 on Aug 8
- #664 on Aug 9
- #654 on Aug 10
- #628 on Aug 11
- #601 on Aug 12
- #596 on Aug 13
- #577 on Aug 14
- #547 on Aug 15
- #534 on Aug 16
- #519 on Aug 17
- #528 on Aug 18
- #516 on Aug 19
- #510 on Aug 20
- #509 on Aug 21
- #490 on Aug 22
- #508 on Aug 23
- #502 on Aug 24
- #505 on Aug 25
- #494 on Aug 26
- #501 on Aug 27
- #494 on Aug 28
- #478 on Aug 29
- #452 on Aug 30
- #467 on Aug 31
- #468 on Sep 1
- #461 on Sep 2
- #463 on Sep 3
- #447 on Sep 4
- #446 on Sep 5
- #441 on Sep 6
- #438 on Sep 7
- #444 on Sep 8
- #434 on Sep 9
- #430 on Sep 10
- #427 on Sep 11
- #421 on Sep 12
- #418 on Sep 13
- #423 from Sep 14 to Sep 15
- #425 on Sep 16
- #418 on Sep 17
- #423 on Sep 18
- #412 on Sep 19
- #418 on Sep 20
- #415 on Sep 21
- #422 on Sep 22
- #412 on Sep 23
- #405 on Sep 24
- #403 on Sep 25
- #400 on Sep 26
- #410 on Sep 27
- #415 on Sep 28
- #424 on Sep 29
- #417 on Sep 30
- #414 on Oct 1
- #416 on Oct 2
- #402 on Oct 3
- #400 on Oct 4
- #408 on Oct 5
- #411 on Oct 6
- #414 on Oct 7
- #422 on Oct 8
- #430 on Oct 9
- #417 on Oct 10
- #419 on Oct 11
- #426 on Oct 12
- #425 on Oct 13
- #424 on Oct 14
- #427 on Oct 15
- #431 on Oct 16
- #422 on Oct 17
- #425 on Oct 18
- #429 on Oct 19
- #431 on Oct 20
- #434 on Oct 21
- #433 from Oct 22 to Oct 23
- #427 on Oct 24
- #449 on Oct 25
Before you install
Local scriptsAutomated signals, not a guarantee
Checked Nov 3, 2026
Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.
Show 49 notes
skills/agy-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/agy-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/aider-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/aider-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/aider-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/cline-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/cline-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/codex-background/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/codex-background/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/codex-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/codex-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/commandcode-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/commandcode-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/copilot-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/copilot-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/cursor-delegate/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- No broad tool permissions
- No dangerous shell commands
- 61 network referencesin 21 files: codex-background/references/configuration.md:272 agy-delegate/SKILL.md:143 aider-delegate/SKILL.md:50 and 18 more
- No hidden or obfuscated text
- 94 instruction override signalsin 9 files: agy-delegate/scripts/relay.mjs:23 claude-delegate/scripts/relay.mjs:72 agy-delegate/references/dispatch-and-poll.md:39 and 6 more
- 28 executable scriptsin 28 files: agy-delegate/scripts/relay.mjs:1 aider-delegate/scripts/relay.mjs:1 claude-delegate/scripts/relay.mjs:1 and 25 more
- 37 references to secretsin 23 files: aider-delegate/SKILL.md:14 delegate-setup/references/setup-dialogue.md:96 muse-delegate/SKILL.md:13 and 20 more
- No requests to vote in Hot
Show all 220 signals by file
skills/agy-delegate/SKILL.md
- Instruction override signalline 130
Antigravity owns its own permission policy. The relay does not bypass it by default. Use `--dangerously-skip-permissions` only when the human explicitly accepts that Antigravity may - Instruction override signalline 131
`--dangerously-skip-permissions` only when the human explicitly accepts that Antigravity may - Instruction override signalline 132
auto-approve tool permission requests. `--read-only` composes `--sandbox` with `--dangerously-skip-permissions`: the sandbox is the enforcement — writes inside the workspace are overlaid and discarded, and paths outside it fail with EPERM — - Instruction override signalline 133
`--dangerously-skip-permissions`: the sandbox is the enforcement — writes inside the workspace - Instruction override signalline 136
`--dangerously-skip-permissions`, which alone (without the sandbox) is full access. Use - Network accessline 143
exact-match behavior: on Windows with agy 1.2.5, `command(<name>)` matched only a bare command with no arguments — `command(git)` never allowed a real `git status`, and `command(regex:git .+)` was required ([issue #614 comment](https://gith - Network accessline 144
no arguments — `command(git)` never allowed a real `git status`, and `command(regex:git .+)` was required ([issue #614 comment](https://github.com/google-antigravity/antigravity-cli/issues/614#issuecomment-5466617752)). On macOS with agy 1. - Network accessline 145
required ([issue #614 comment](https://github.com/google-antigravity/antigravity-cli/issues/614#issuecomment-5466617752)). - Network accessline 147
exact-match trap is not a constant across versions. Second, on Windows, an open upstream defect in Antigravity's permission engine ([issue #614](https://github.com/google-antigravity/antigravity-cli/issues/614)) splits resolved paths - Network accessline 148
upstream defect in Antigravity's permission engine ([issue #614](https://github.com/google-antigravity/antigravity-cli/issues/614)) splits resolved paths on whitespace, so a binary under `C:\Program Files\...` — which is where `git` itself - Network accessline 149
([issue #614](https://github.com/google-antigravity/antigravity-cli/issues/614)) splits resolved paths
skills/agy-delegate/references/dispatch-and-poll.md
- Instruction override signalline 39
| `--conversation <id>` | Continue a specific Antigravity conversation; send only the delta brief. | | `--sandbox` | Enable Antigravity's terminal sandbox for the run. | | `--read-only` | Run under agy's sandbox with tool approval auto-gran - Instruction override signalline 40
| `--sandbox` | Enable Antigravity's terminal sandbox for the run. | | `--read-only` | Run under agy's sandbox with tool approval auto-granted inside it (`--sandbox --dangerously-skip-permissions`); reads and writes are both confined to the - Instruction override signalline 41
| `--read-only` | Run under agy's sandbox with tool approval auto-granted inside it (`--sandbox --dangerously-skip-permissions`); reads and writes are both confined to the workspace — a write appears to succeed to the agent but is overlaid - Instruction override signalline 42
| `--dangerously-skip-permissions` | Pass Antigravity's permission-bypass flag; mutually exclusive with `--read-only`. Never use this unless the human explicitly accepts it. | - Instruction override signalline 68
log, and stderr capture - `workdir`, `model`, `effort`, `stallTimeout` (the `--stall-timeout` you passed, or `null`), `project` (the `--project` you passed, vs `projectId` parsed from the log), `sandbox`, `readOnly`, `dangerouslySkipPermi - Instruction override signalline 69
- `workdir`, `model`, `effort`, `stallTimeout` (the `--stall-timeout` you passed, or `null`), `project` (the `--project` you passed, vs `projectId` parsed from the log), `sandbox`, `readOnly`, `dangerouslySkipPermissions`, `resumed` (true - Instruction override signalline 70
`sandbox`, `readOnly`, `dangerouslySkipPermissions`, `resumed` (true for a `--resume-last` or `--conversation` - Instruction override signalline 124
was insufficient on Windows/agy 1.2.5 but did work on macOS/agy 1.2.0 — so measure before assuming; see [Permission engine traps](#permission-engine-traps) below before re-dispatching or asking to use `--dangerously-skip-permissions`. - Instruction override signalline 125
see [Permission engine traps](#permission-engine-traps) below before re-dispatching or asking to use `--dangerously-skip-permissions`. - **Empty `finalMessage`:** a run with edits may still be correct - check `touchedFiles`, the diff, and - Instruction override signalline 126
`--dangerously-skip-permissions`. - Network accessline 167
first thing to check on *any* denial, before suspecting the Windows bug below: use `command(regex:git .+)` (or a narrower pattern for a specific subcommand) to match real invocations. Source: [issue #614 comment](https://github.com/google-a - Network accessline 169
Source: [issue #614 comment](https://github.com/google-antigravity/antigravity-cli/issues/614#issuecomment-5466617752). - Instruction override signalline 216
Windows path-splitting bug too. `command(*)` (documented in issue #614) is the broader fallback if a targeted regex still doesn't match - it approves all command execution, so treat it the same as `--dangerously-skip-permissions` for th - Instruction override signalline 217
targeted regex still doesn't match - it approves all command execution, so treat it the same as `--dangerously-skip-permissions` for that run. - **Directory permissions:** Write directory paths literally without a wildcard, e.g. `write_fi - Instruction override signalline 218
`--dangerously-skip-permissions` for that run. - Instruction override signalline 221
If settings allow-rules cannot resolve the denial, ask the human before re-dispatching with `--dangerously-skip-permissions`;
skills/agy-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node - Instruction override signalline 23
* * Antigravity owns its own permission policy. This helper does not pass * --dangerously-skip-permissions by default; opt into that flag only when the - Instruction override signalline 24
* Antigravity owns its own permission policy. This helper does not pass * --dangerously-skip-permissions by default; opt into that flag only when the * human explicitly accepts it. Pass --sandbox to enable Antigravity's terminal - Instruction override signalline 25
* --dangerously-skip-permissions by default; opt into that flag only when the - Instruction override signalline 27
* sandbox for the run. * * `--read-only` composes both: `--sandbox --dangerously-skip-permissions`. That - Instruction override signalline 28
* * `--read-only` composes both: `--sandbox --dangerously-skip-permissions`. That * pairing looks alarming and is deliberate — the sandbox is the enforcement and - Instruction override signalline 29
* `--read-only` composes both: `--sandbox --dangerously-skip-permissions`. That - Instruction override signalline 63
* --read-only Review/diagnosis with no edits: runs under agy's sandbox with * tool approval auto-granted inside it (`--sandbox * --dangerously-skip-permissions`). Reads and - Instruction override signalline 64
* tool approval auto-granted inside it (`--sandbox * --dangerously-skip-permissions`). Reads and writes are both * confined to the workspace. Mutually exclusive - Instruction override signalline 65
* --dangerously-skip-permissions`). Reads and writes are both - Instruction override signalline 66
* confined to the workspace. Mutually exclusive with * --dangerously-skip-permissions as a flag: that alone, without * the sandbox, is full access. - Instruction override signalline 67
* --dangerously-skip-permissions as a flag: that alone, without - Instruction override signalline 68
* the sandbox, is full access. * --dangerously-skip-permissions * Auto-approve Antigravity tool permission requests. Use only with human approval. - Instruction override signalline 69
* --dangerously-skip-permissions - Instruction override signalline 153
if (field === "permissionMode" && (flagged.has("permissionMode") || flagged.has("readOnly"))) continue; if (field === "planOnly" && (flagged.has("planOnly") || flagged.has("readOnly"))) continue; if (field === "readOnly" && (flagged - Instruction override signalline 154
if (field === "planOnly" && (flagged.has("planOnly") || flagged.has("readOnly"))) continue; if (field === "readOnly" && (flagged.has("readOnly") || flagged.has("dangerouslySkipPermissions"))) continue; if (field === "force" && flagg - Instruction override signalline 155
if (field === "readOnly" && (flagged.has("readOnly") || flagged.has("dangerouslySkipPermissions"))) continue; - Instruction override signalline 211
case "--sandbox": opts.sandbox = true; flagged.add("sandbox"); break; case "--read-only": opts.readOnly = true; flagged.add("readOnly"); break; case "--dangerously-skip-permissions": - Instruction override signalline 212
case "--read-only": opts.readOnly = true; flagged.add("readOnly"); break; case "--dangerously-skip-permissions": opts.dangerouslySkipPermissions = true; - Instruction override signalline 213
case "--dangerously-skip-permissions": - Instruction override signalline 214
opts.dangerouslySkipPermissions = true; - Instruction override signalline 215
flagged.add("dangerouslySkipPermissions"); - Instruction override signalline 228
fail(`invalid --effort "${opts.effort}" (expected: low, medium, high)`); } if (opts.readOnly && opts.dangerouslySkipPermissions) { - Instruction override signalline 229
} if (opts.readOnly && opts.dangerouslySkipPermissions) { fail("--read-only and --dangerously-skip-permissions are mutually exclusive; pass only one"); - Instruction override signalline 230
if (opts.readOnly && opts.dangerouslySkipPermissions) { - Instruction override signalline 231
fail("--read-only and --dangerously-skip-permissions are mutually exclusive; pass only one"); - Instruction override signalline 486
argv.push("--sandbox", "--dangerously-skip-permissions"); - Instruction override signalline 487
} else { if (opts.sandbox) argv.push("--sandbox"); if (opts.dangerouslySkipPermissions) argv.push("--dangerously-skip-permissions"); - Instruction override signalline 488
if (opts.sandbox) argv.push("--sandbox"); if (opts.dangerouslySkipPermissions) argv.push("--dangerously-skip-permissions"); } - Instruction override signalline 489
if (opts.dangerouslySkipPermissions) argv.push("--dangerously-skip-permissions"); - Instruction override signalline 577
readOnly: opts.readOnly, readOnlyViolation: null, dangerouslySkipPermissions: opts.dangerouslySkipPermissions, - Instruction override signalline 578
readOnlyViolation: null, dangerouslySkipPermissions: opts.dangerouslySkipPermissions, resumed: Boolean(opts.resumeLast || opts.conversation), - Instruction override signalline 579
dangerouslySkipPermissions: opts.dangerouslySkipPermissions, - Instruction override signalline 834
} : permissionDenied ? { error: `Antigravity auto-denied the ${permissionDenied[1]} permission because headless --print cannot prompt; ask the human whether to re-dispatch with --dangerously-skip-permissions and treat that - Instruction override signalline 835
: permissionDenied ? { error: `Antigravity auto-denied the ${permissionDenied[1]} permission because headless --print cannot prompt; ask the human whether to re-dispatch with --dangerously-skip-permissions and treat that run as fu - Instruction override signalline 836
? { error: `Antigravity auto-denied the ${permissionDenied[1]} permission because headless --print cannot prompt; ask the human whether to re-dispatch with --dangerously-skip-permissions and treat that run as full access` }
skills/aider-delegate/SKILL.md
- Access to secretsline 14
[Credential-related source omitted] - Network accessline 50
1. Install Aider - `python -m pip install aider-chat`, or the standalone installer from the [Aider install docs](https://aider.chat/docs/install.html). - Network accessline 51
[Aider install docs](https://aider.chat/docs/install.html). - Network accessline 52
[Credential-related source omitted] - Access to secretsline 52
[Credential-related source omitted] - Network accessline 53
[Credential-related source omitted] - Access to secretsline 53
[Credential-related source omitted] - Network accessline 68
node "<skill-dir>/scripts/relay.mjs" --brief brief.txt --cd /path/to/repo \ --model openai/<served-model-name> --api-base http://127.0.0.1:<port>/v1 - Network accessline 69
--model openai/<served-model-name> --api-base http://127.0.0.1:<port>/v1 - Access to secretsline 76
[Credential-related source omitted] - Access to secretsline 89
[Credential-related source omitted]
skills/aider-delegate/references/dispatch-and-poll.md
- Network accessline 50
node "<skill-dir>/scripts/relay.mjs" --brief brief.txt --cd /path/to/repo \ --model openai/<served-model-name> --api-base http://127.0.0.1:<port>/v1 \ - Network accessline 51
--model openai/<served-model-name> --api-base http://127.0.0.1:<port>/v1 \ - Access to secretsline 56
[Credential-related source omitted]
skills/aider-delegate/references/writing-the-brief.md
- Access to secretsline 13
[Credential-related source omitted]
skills/aider-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node - Network accessline 468
}); printSummary(result, resultPath); process.stderr.write("relay: `aider` not found on PATH. Install Aider (https://aider.chat/docs/install.html) and configure a model.\n"); - Network accessline 470
process.stderr.write("relay: `aider` not found on PATH. Install Aider (https://aider.chat/docs/install.html) and configure a model.\n"); - Access to secretsline 503
[Credential-related source omitted] - Access to secretsline 505
[Credential-related source omitted]
skills/claude-delegate/SKILL.md
skills/claude-delegate/references/dispatch-and-poll.md
- Instruction override signalline 43
| `--dangerously-skip-permissions` | Opt into Claude's `bypassPermissions`; mutually exclusive with `--read-only`. | - Access to secretsline 71
[Credential-related source omitted] - Instruction override signalline 130
`--dangerously-skip-permissions` passes Claude's flag and records - Instruction override signalline 171
Run metadata includes `workdir`, `model`, `effort`, `maxTurns`, `maxBudgetUsd`, optional `autocompact`, `timeout`, `readOnly`, `resumed`, `resumeLast`, `toolSurface`, `shellSandbox`, `dangerouslySkipPermissions`, timestamps, and - Instruction override signalline 172
`autocompact`, `timeout`, `readOnly`, `resumed`, `resumeLast`, `toolSurface`, `shellSandbox`, `dangerouslySkipPermissions`, timestamps, and all artifact paths. Failed, timed-out, and aborted runs include `stderrTail` when available; - Instruction override signalline 173
`resumed`, `resumeLast`, `toolSurface`, `shellSandbox`, `dangerouslySkipPermissions`, timestamps, and
skills/claude-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node - Instruction override signalline 72
* --dangerously-skip-permissions Opt into Claude's bypassPermissions mode. - Instruction override signalline 229
if ( field === "readOnly" && (flagged.has("readOnly") || flagged.has("dangerouslySkipPermissions")) - Instruction override signalline 230
field === "readOnly" && (flagged.has("readOnly") || flagged.has("dangerouslySkipPermissions")) ) { - Instruction override signalline 231
(flagged.has("readOnly") || flagged.has("dangerouslySkipPermissions")) - Instruction override signalline 307
case "--session": opts.session = next(); break; case "--read-only": opts.readOnly = true; flagged.add("readOnly"); break; case "--dangerously-skip-permissions": - Instruction override signalline 308
case "--read-only": opts.readOnly = true; flagged.add("readOnly"); break; case "--dangerously-skip-permissions": opts.dangerouslySkipPermissions = true; - Instruction override signalline 309
case "--dangerously-skip-permissions": - Instruction override signalline 310
opts.dangerouslySkipPermissions = true; - Instruction override signalline 311
flagged.add("dangerouslySkipPermissions"); - Instruction override signalline 320
fail("--resume-last and --session are mutually exclusive; pass only one"); } if (opts.readOnly && opts.dangerouslySkipPermissions) { - Instruction override signalline 321
} if (opts.readOnly && opts.dangerouslySkipPermissions) { fail("--read-only and --dangerously-skip-permissions are mutually exclusive"); - Instruction override signalline 322
if (opts.readOnly && opts.dangerouslySkipPermissions) { - Instruction override signalline 323
fail("--read-only and --dangerously-skip-permissions are mutually exclusive"); - Instruction override signalline 500
function selectedPermissionMode(opts) { if (opts.readOnly) return "plan"; if (opts.dangerouslySkipPermissions) return "bypassPermissions"; - Instruction override signalline 501
if (opts.readOnly) return "plan"; if (opts.dangerouslySkipPermissions) return "bypassPermissions"; return "acceptEdits"; - Instruction override signalline 502
if (opts.dangerouslySkipPermissions) return "bypassPermissions"; - Instruction override signalline 595
if (opts.readOnly) { argv.push("--permission-mode", "plan"); } else if (opts.dangerouslySkipPermissions) { - Instruction override signalline 596
argv.push("--permission-mode", "plan"); } else if (opts.dangerouslySkipPermissions) { argv.push("--dangerously-skip-permissions"); - Instruction override signalline 597
} else if (opts.dangerouslySkipPermissions) { - Instruction override signalline 598
argv.push("--dangerously-skip-permissions"); - Instruction override signalline 961
? "unsupported-on-native-windows" : "strict", dangerouslySkipPermissions: opts.dangerouslySkipPermissions, - Instruction override signalline 962
: "strict", dangerouslySkipPermissions: opts.dangerouslySkipPermissions, claudeVersion: version, - Instruction override signalline 963
dangerouslySkipPermissions: opts.dangerouslySkipPermissions, - Instruction override signalline 1205
); lines.push(`permission mode: ${result.permissionMode} · shell sandbox: ${result.shellSandbox}`); if (result.dangerouslySkipPermissions) { - Instruction override signalline 1206
lines.push(`permission mode: ${result.permissionMode} · shell sandbox: ${result.shellSandbox}`); if (result.dangerouslySkipPermissions) { lines.push("warning: bypassPermissions was explicitly enabled; direct file tools can reach bey - Instruction override signalline 1207
if (result.dangerouslySkipPermissions) {
skills/cline-delegate/SKILL.md
- Access to secretsline 36
[Credential-related source omitted]
skills/cline-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/codex-background/references/configuration.md
- Network accessline 272
RPC through an early-yielding exec wrapper: that recreates the polling boundary. The host requires tool_timeout_sec above the declared relay budget; required=true makes MCP initialization failure explicit. Current documentation: [App Server - Network accessline 272
RPC through an early-yielding exec wrapper: that recreates the polling boundary. The host requires tool_timeout_sec above the declared relay budget; required=true makes MCP initialization failure explicit. Current documentation: [App Server - Network accessline 273
requires tool_timeout_sec above the declared relay budget; required=true makes MCP initialization failure explicit. Current documentation: [App Server](https://developers.openai.com/codex/app-server) and [configuration reference](https://de - Network accessline 273
requires tool_timeout_sec above the declared relay budget; required=true makes MCP initialization failure explicit. Current documentation: [App Server](https://developers.openai.com/codex/app-server) and [configuration reference](https://de - Network accessline 274
failure explicit. Current documentation: [App Server](https://developers.openai.com/codex/app-server) - Network accessline 274
failure explicit. Current documentation: [App Server](https://developers.openai.com/codex/app-server) - Network accessline 275
and [configuration reference](https://developers.openai.com/codex/config-reference). - Network accessline 275
and [configuration reference](https://developers.openai.com/codex/config-reference).
skills/codex-background/scripts/bootstrap.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/codex-background/scripts/bootstrap.test.mjs
- Executable scriptline 1
Executable file
skills/codex-background/scripts/server.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/codex-background/scripts/server.test.mjs
- Executable scriptline 1
Executable file
skills/codex-background/scripts/supervisor.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/codex-background/scripts/windows-job.ps1
- Executable scriptline 1
Executable file
skills/codex-delegate/references/dispatch-and-poll.md
skills/codex-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/commandcode-delegate/SKILL.md
- Instruction override signalline 40
refused by the CLI's permission layer, and headless mode has no prompt to grant them mid-run. This is the relay's `--read-only`. - **`--yolo` (alias `--dangerously-skip-permissions`):** every tool is allowed, anywhere the process - Instruction override signalline 41
is the relay's `--read-only`. - **`--yolo` (alias `--dangerously-skip-permissions`):** every tool is allowed, anywhere the process can reach. There is no filesystem sandbox and no path restriction. This is what an implementation - Instruction override signalline 42
- **`--yolo` (alias `--dangerously-skip-permissions`):** every tool is allowed, anywhere the process
skills/commandcode-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node - Instruction override signalline 20
* and glob but every write is refused by its permission layer. With `--yolo * (alias `--dangerously-skip-permissions`) every tool is allowed, everywhere the - Instruction override signalline 21
* (alias `--dangerously-skip-permissions`) every tool is allowed, everywhere the
skills/copilot-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/cursor-delegate/SKILL.md
- Network accessline 33
1. `cursor-agent --version` succeeds. If not, follow the installer for your platform at [cursor.com/cli](https://cursor.com/cli), inspect what it will run, and authenticate with - Network accessline 34
[cursor.com/cli](https://cursor.com/cli), inspect what it will run, and authenticate with
skills/cursor-delegate/references/dispatch-and-poll.md
- Network accessline 14
Follow the installer for your platform at [cursor.com/cli](https://cursor.com/cli), inspect what it
skills/cursor-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node - Network accessline 477
}); printSummary(result, resultPath); process.stderr.write("relay: `cursor-agent` not found on PATH. Install the Cursor CLI (https://cursor.com/cli) and run `cursor-agent login`.\n"); - Network accessline 478
printSummary(result, resultPath); process.stderr.write("relay: `cursor-agent` not found on PATH. Install the Cursor CLI (https://cursor.com/cli) and run `cursor-agent login`.\n"); process.exit(127); - Network accessline 479
process.stderr.write("relay: `cursor-agent` not found on PATH. Install the Cursor CLI (https://cursor.com/cli) and run `cursor-agent login`.\n");
skills/delegate-setup/references/setup-dialogue.md
skills/delegate-setup/scripts/config.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/delegate-setup/scripts/discover.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/delegate-setup/scripts/implementers.mjs
skills/delegate-setup/scripts/lane.mjs
- Executable scriptline 1
#!/usr/bin/env node - Instruction override signalline 140
field === "readOnly" && (flagged.has("readOnly") || flagged.has("dangerouslySkipPermissions") || - Instruction override signalline 141
(flagged.has("readOnly") || flagged.has("dangerouslySkipPermissions") || flagged.has("fullAccess")) - Instruction override signalline 142
flagged.has("dangerouslySkipPermissions") ||
skills/grok-delegate/SKILL.md
skills/grok-delegate/references/dispatch-and-poll.md
- Access to secretsline 15
[Credential-related source omitted]
skills/grok-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/kimi-delegate/SKILL.md
- Network accessline 31
1. Install Kimi Code with `brew install kimi-code` on macOS/Linux, or use the native installer from the [official Kimi Code documentation](https://moonshotai.github.io/kimi-code/en/). - Network accessline 32
the [official Kimi Code documentation](https://moonshotai.github.io/kimi-code/en/).
skills/kimi-delegate/references/dispatch-and-poll.md
- Network accessline 14
Install with `brew install kimi-code` on macOS/Linux or use a native installer from the [official Kimi Code documentation](https://moonshotai.github.io/kimi-code/en/). `kimi login` uses a - Network accessline 15
[official Kimi Code documentation](https://moonshotai.github.io/kimi-code/en/). `kimi login` uses a
skills/kimi-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/muse-delegate/SKILL.md
skills/muse-delegate/references/dispatch-and-poll.md
- Access to secretsline 16
[Credential-related source omitted]
skills/muse-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/omp-delegate/SKILL.md
skills/omp-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/opencode-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/pi-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node
skills/qoder-delegate/SKILL.md
- Network accessline 35
If the binary is missing, install it from Qoder's [official Quick Start](https://docs.qoder.com/en/cli/quick-start). Authenticate with `qodercli login`, - Network accessline 36
[official Quick Start](https://docs.qoder.com/en/cli/quick-start). Authenticate with `qodercli login`, - Access to secretsline 37
[Credential-related source omitted]
skills/qoder-delegate/references/dispatch-and-poll.md
- Network accessline 14
Install and authenticate using Qoder's [official Quick Start](https://docs.qoder.com/en/cli/quick-start). Use `qodercli login` interactively or - Network accessline 15
[official Quick Start](https://docs.qoder.com/en/cli/quick-start). Use `qodercli login` interactively or - Access to secretsline 16
[Credential-related source omitted]
skills/qoder-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node - Network accessline 415
}); printSummary(result, resultPath); process.stderr.write("relay: `qodercli` not found on PATH. Install from https://docs.qoder.com/en/cli/quick-start, then run `qodercli login` or set QODER_PERSONAL_ACCESS_TOKEN for automation.\n"); - Network accessline 416
printSummary(result, resultPath); process.stderr.write("relay: `qodercli` not found on PATH. Install from https://docs.qoder.com/en/cli/quick-start, then run `qodercli login` or set QODER_PERSONAL_ACCESS_TOKEN for automation.\n"); proce - Network accessline 417
[Credential-related source omitted] - Access to secretsline 417
[Credential-related source omitted]
skills/vibe-delegate/SKILL.md
skills/vibe-delegate/references/dispatch-and-poll.md
- Network accessline 13
Install [`uv`](https://docs.astral.sh/uv/getting-started/installation/), then install Vibe with - Access to secretsline 18
[Credential-related source omitted] - Network accessline 21
Upstream Vibe works on Windows but officially supports and targets UNIX. This repository has not smoke-tested the relay's native Windows launch; consult the [official Mistral Vibe documentation](https://github.com/mistralai/mistral-vibe) fo - Network accessline 23
[official Mistral Vibe documentation](https://github.com/mistralai/mistral-vibe) for Windows guidance. - Network accessline 99
- **`status: "vibe_unavailable"` (exit 127):** `vibe` isn't on PATH. Install [`uv`](https://docs.astral.sh/uv/getting-started/installation/), run - Network accessline 100
[`uv`](https://docs.astral.sh/uv/getting-started/installation/), run - Access to secretsline 101
[Credential-related source omitted]
skills/vibe-delegate/scripts/relay.mjs
skills/warp-delegate/SKILL.md
skills/warp-delegate/references/dispatch-and-poll.md
- Network accessline 65
"resumed": false, "runId": "00000000-0000-0000-0000-000000000000", "runUrl": "https://oz.warp.dev/runs/00000000-0000-0000-0000-000000000000", - Network accessline 66
"runId": "00000000-0000-0000-0000-000000000000", "runUrl": "https://oz.warp.dev/runs/00000000-0000-0000-0000-000000000000", "conversationId": null, - Network accessline 67
"runUrl": "https://oz.warp.dev/runs/00000000-0000-0000-0000-000000000000",
skills/warp-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node - Network accessline 535
}); printSummary(result, resultPath); process.stderr.write("relay: `oz` not found on PATH. Install the Warp Agent CLI (https://docs.warp.dev/cli/), then authenticate with `oz login` (or set WARP_API_KEY on a headless host).\n"); - Network accessline 536
printSummary(result, resultPath); process.stderr.write("relay: `oz` not found on PATH. Install the Warp Agent CLI (https://docs.warp.dev/cli/), then authenticate with `oz login` (or set WARP_API_KEY on a headless host).\n"); process.exi - Network accessline 537
[Credential-related source omitted] - Access to secretsline 537
[Credential-related source omitted]
skills/zcode-delegate/SKILL.md
- Network accessline 60
"zai": { "kind": "anthropic", "options": { "apiKeyRequired": true, "baseURL": "https://api.z.ai/api/anthropic" }, - Network accessline 61
"kind": "anthropic", "options": { "apiKeyRequired": true, "baseURL": "https://api.z.ai/api/anthropic" }, "models": { "glm-5.1": { "name": "GLM-5.1" } } - Network accessline 62
"options": { "apiKeyRequired": true, "baseURL": "https://api.z.ai/api/anthropic" }, - Access to secretsline 71
[Credential-related source omitted]
skills/zcode-delegate/scripts/relay.mjs
- Executable scriptline 1
#!/usr/bin/env node
How the signals fit together
Facts from a static read of the files at the pinned commit. They describe what the code does, not why.
Some files could not be read; the facts below still hold.
Signal confidence: 220 facts
- Instruction override signal in prose of the SKILL.mdagy-delegate/SKILL.md:130High confidence
- Instruction override signal in prose of the SKILL.mdagy-delegate/SKILL.md:131High confidence
- Instruction override signal in prose of the SKILL.mdagy-delegate/SKILL.md:132High confidence
- Instruction override signal in prose of the SKILL.mdagy-delegate/SKILL.md:133High confidence
- Instruction override signal in prose of the SKILL.mdagy-delegate/SKILL.md:136High confidence
- Network access in prose of the SKILL.mdagy-delegate/SKILL.md:143Low confidence
- Network access in prose of the SKILL.mdagy-delegate/SKILL.md:144Low confidence
- Network access in prose of the SKILL.mdagy-delegate/SKILL.md:145Low confidence
- Network access in prose of the SKILL.mdagy-delegate/SKILL.md:147Low confidence
- Network access in prose of the SKILL.mdagy-delegate/SKILL.md:148Low confidence
- Network access in prose of the SKILL.mdagy-delegate/SKILL.md:149Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:39Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:40Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:41Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:42Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:68Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:69Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:70Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:124Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:125Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:126Low confidence
- Network access in prose of documentationagy-delegate/references/dispatch-and-poll.md:167Low confidence
- Network access in prose of documentationagy-delegate/references/dispatch-and-poll.md:169Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:216Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:217Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:218Low confidence
- Instruction override signal in prose of documentationagy-delegate/references/dispatch-and-poll.md:221Low confidence
- Executable script in code of a file that runsagy-delegate/scripts/relay.mjs:1High confidence
- Instruction override signal in a comment of a file that runsagy-delegate/scripts/relay.mjs:23Low confidence
- Instruction override signal in a comment of a file that runsagy-delegate/scripts/relay.mjs:24Low confidence
- Instruction override signal in a comment of a file that runsagy-delegate/scripts/relay.mjs:25Low confidence
- Instruction override signal in a comment of a file that runsagy-delegate/scripts/relay.mjs:27Low confidence
- Instruction override signal in a comment of a file that runsagy-delegate/scripts/relay.mjs:28Low confidence
- Instruction override signal in a comment of a file that runsagy-delegate/scripts/relay.mjs:29Low confidence
- Instruction override signal in a comment of a file that runsagy-delegate/scripts/relay.mjs:63Low confidence
- Instruction override signal in a comment of a file that runsagy-delegate/scripts/relay.mjs:64Low confidence
- Instruction override signal in a comment of a file that runsagy-delegate/scripts/relay.mjs:65Low confidence
- Instruction override signal in a comment of a file that runsagy-delegate/scripts/relay.mjs:66Low confidence
- Instruction override signal in a comment of a file that runsagy-delegate/scripts/relay.mjs:67Low confidence
- Instruction override signal in a comment of a file that runsagy-delegate/scripts/relay.mjs:68Low confidence
- Instruction override signal in a comment of a file that runsagy-delegate/scripts/relay.mjs:69Low confidence
- Instruction override signal in code of a file that runsagy-delegate/scripts/relay.mjs:153Low confidence
- Instruction override signal in code of a file that runsagy-delegate/scripts/relay.mjs:154Low confidence
- Instruction override signal in code of a file that runsagy-delegate/scripts/relay.mjs:155Low confidence
- Instruction override signal in code of a file that runsagy-delegate/scripts/relay.mjs:211Low confidence
- Instruction override signal in code of a file that runsagy-delegate/scripts/relay.mjs:212Low confidence
- Instruction override signal in code of a file that runsagy-delegate/scripts/relay.mjs:213Low confidence
- Instruction override signal in code of a file that runsagy-delegate/scripts/relay.mjs:214Low confidence
- Instruction override signal in code of a file that runsagy-delegate/scripts/relay.mjs:215Low confidence
- Instruction override signal in code of a file that runsagy-delegate/scripts/relay.mjs:228Low confidence
and 170 more
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Compatible | Standard SKILL.md skills/agy-delegate/SKILL.md | .claude/skills/<skill-name>/ |
| Codex | Native | Manifest AGENTS.md | .agents/skills/<skill-name>/ |
| Cursor | Native | Path .cursor/environment.json | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md skills/agy-delegate/SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md skills/agy-delegate/SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md skills/agy-delegate/SKILL.md | .opencode/skills/<skill-name>/ |
- Native The repository ships a manifest, folder or frontmatter field for this agent.
- Compatible A standard SKILL.md that this agent can read.
- Unknown No evidence either way.
Derived automatically from paths, manifests and frontmatter. Not tested on each agent.
Explore alternatives and comparisons
Add this badge to your README
It updates every night and links to this page.
[](https://hotskills.dev/skill/amelnagdy-delegate-skills/)